ABSTRACT. cause this view is based solely on subjective evaluations

Transcription

1 AMERICAN MUSEUM Novltates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y Number 3034, 16 pp., 4 figures February 25, 1992 Phylogeny of Plecotine Bats (Chiroptera: "Vespertilionidae"): Summary of the Evidence and Proposal of a Logically Consistent Taxonomy DARREL R. FROST' AND ROBERT M. TIMM2 Using standard phylogenetic techniques, 25 transformation series of morphological characters and 11 of karyological characters are evaluated in an attempt to recover the phylogenetic history of plecotine vespertilionid bats. Plecotini contains four genera in the topology (Euderma (Barbastella (Plecotus Corynorhinus))). The Plecotini of Hill and Harrison (1987), including Rhogeessa, Baeodon, Nycticeius, and Otonycteris, is rejected be- ABSTRACT cause this view is based solely on subjective evaluations ofbacular overall similarity, and is clearly in disagreement with other lines of evidence from anatomy and karyology. Idionycteris is synonymized with Euderma because L phyllotis and E. maculatum are sister species. Corynorhinus is removed from the synonymy of Plecotus. The relationships within Corynorhinus and Plecotus are not resolved. INTRODUCTION Plecotine bats are a small group of largeeared bats found in temperate North America four (Williams et al., 1970; Honacki et al., ker et al., 1974; Koopman, 1984a, 1984b), and Europe. The ten species of these vespertilionid bats have been placed implicitly or 1942) genera. Although it differs from the 1982; Hill and Harrison, 1987), or five (Tate, explicitly either in three (Handley, 1959; Ba- most recent revisions, for purposes of this ' Assistant Curator, Department of Herpetology and Ichthyology, American Museum of Natural History. 2 Curator, Division of Mammals, Museum of Natural History; Associate Professor, Department of Systematics and Ecology, The University of Kansas, Lawrence, KS Copyright American Museum of Natural History 1992 ISSN / Price $2.00

2 2 AMERICAN MUSEUM NOVITATES NO study we will use the five-genus taxonomy: Barbastella (B. barbastella, B. leucomelas), Euderma (E. maculatum), Idionycteris (L phyllotis), Plecotus (P. austriacus, P. auritus, and P. teneriffae), and Corynorhinus (C. mexicanus, C. rafinesquii, and C. townsendii). The term "plecotine" and its formal equivalents have had a history of varied use. Dobson (1875) recognized the Group Plecoti for "Vespertilionidae"3 having large ears and either rudimentary noseleaves or grooves on the dorsal surface of the muzzle, and Miller (1897) used the name Plecotinae for the American bats of the genera Corynorhinus and Euderma. Subsequently, however, most authors have either abandoned the use of the formal taxonomic name (e.g., Miller, 1907; Simpson, 1945; Handley, 1959; Sokolov, 1973) or have used it for a tribe, Plecotini, of varying generic content within the "Vespertilioninae" (e.g., Tate, 1942; Baker et al., 1974; Koopman, 1984a, 1984b; Hill and Harrison, 1987). Williams et al. (1970) recognized a Group Plecoti composed of the plecotines in the sense that we use the term. With variation in the number of genera recognized and with the exception of the recent arrangement ofhill and Harrison (1987) (see discussion below) this is generally how the term is currently used. Handley (1959) provided a wealth of descriptive information on the group and posited a phylogeny of (Barbastella (Euderma (Idionycteris (Corynorhinus Plecotus)))). Williams et al. (1970) suggested, on the basis of karyology, the phylogeny (Barbastella ((Euderma Idionycteris) (Plecotus Corynorhinus))). The evidence provided by Williams et al. (1970), supporting a special relationship of Plecotus and Corynorhinus, was solely overall similarity of morphology and karyotype, although the evidence relating Euderma and Idionycteris was synapomorphic. This view that Idionycteris is generically distinct from Plecotus (sensu lato as including Plecotus and Corynorhinus) was disputed by Baker et al. (1974), who argued that the Idionycteris karyotype might not be dissimilar enough from the Plecotus karyotype to war- 3We place quotations around taxonomic names that denote "taxa" that are not demonstrably monophyletic (Wiley, 1979). rant generic distinction, particularly without morphological features to support this assertion. Koopman (in Honacki et al., 1982: 177) also disputed the position of Williams et al. (1970) that Idionycteris should be considered generically distinct from Plecotus, although this view was based on evidence provided by Handley (1959). Stock (1983) provided a karyologically based three-taxon cladogram of ((Euderma Idionycteris) Corynorhinus townsendii). More recently, Leniec et al. (1987) summarized the karyology of plecotines and supported the monophyly ofthe group, including Barbastella. They also made the novel suggestion that the short-eared bats of the genus Barbastella are derived from the large-eared plecotines, in the arrangement ((Euderma Idionycteris) (Barbastella Plecotus Corynorhinus)). The objective of this study is to attempt to resolve the alternative phylogenetic hypotheses for plecotines that have been advanced and to suggest a taxonomy of these bats that is logically consistent with recovered phylogeny if previous taxonomies are found to be wanting. ACKNOWLEDGMENTS We thank Charles J. Cole, Nancy Simmons, Lynne Frost, Arnold G. Kluge, Karl F. Koopman, and Charles W. Myers for constructive criticism of the manuscript. Karl Koopman was particularly helpful in discussions of bat morphology and relationships, although this should not be construed necessarily as agreement with our conclusions. Arnold Kluge provided especially interesting criticisms regarding methodology and underlying philosophy. We also thank Michael D. Carleton and Linda K. Gordon (National Museum ofnatural History [USNM]), Bruce D. Patterson (Field Museum of Natural History [FMNH]), and Karl F. Koopman (AMNH) for access to and loan of specimens under their care. This study was funded in part by a grant to Timm from the University of Kansas General Research Fund. EVIDENCE OF MONOPHYLY OF THE PLECOTINES AND CHOICE OF OUTGROUPS Tate (1942) regarded plecotines, Euderma,

3 1 992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 3 Plecotus, Corynorhinus, and Idionycteris, to be a subgroup, Plecotini, of his "Myotini" (sensu lato). The taxa in the "Myotini" (sensu lato), exclusive of the plecotines, he referred to as "Myotini" (sensu stricto), a residual assemblage containing "Myotis" and Lasionycteris. Tate's "Myotini" (sensu lato) is characterized by having reduced P2-3/2-3 -not a particularly helpful character inasmuch as these teeth are absent in some other groups ofvespertilionids (e.g., Antrozous) and might actually be synapomorphic of a larger group. Tate considered Barbastella to be part of the pipistrelloid radiation, and therefore phylogenetically distant from the plecotines. However, Handley (1959) posited that the traits suggested by Tate that would place Barbastella in the pipistrelloid group are plesiomorphies also shared with "Myotis," with the exception of the reduction of the P1 in Barbastella and "Pipistrellus." In in our examination of specimens, reduction of the P1 seems to be highly variable within species and is likely either plesiomorphic or too variable to be useful. Handley (1959), following Miller (1907), considered Barbastella to be the nearest relative of the plecotine group. This view rested largely on the overall similarity ofthe dentition and skull, but Handley noted at least one synapomorphy of this group, zygoma not bowed out posteriorly. Additionally, Barbastella shares with the other plecotines the ability to roll its ears into a rams-horn shape, as evidenced by the arrangement ofcartilaginous supports in the ear (K. Koopman, personal commun.). Another character that may be synapomorphic for the plecotines, although it occurs in many other vespertilionid bats, is the interauricular septum (low in Corynorhinus) that is absent in "Myotis" and other myotines. Bickham (1979) considered the plecotines to be within the "Myotis" group of genera (vespertilionids excluding Miniopterus and the "Eptesicus" group of genera) that share a particular karyotype. Although Bickham (1979) regarded the "Myotis" karyotype as primitive within the family, Stock (1983) and Leniec et al. (1987) have argued that this karyotype is, in fact, derived from the "Eptesicus" karyotype by centric fusion, thereby positing a synapomorphy for the "Myotis" group of genera, including the plecotines. Within the "Myotis" generic group, Bickham (1979) suggested that the plecotine (+ Barbastella, fide Leniec et al., 1987) karyotype was derived by centric fusion (of chromosomes 15 and 11) from the "Myotis" karyotype. Stock (1983) did not examine Barbastella or Plecotus, but did concur with Bickham (1979) that the plecotines are closely related to "Myotis," differing by a single chromosomal fusion. Leniec et al. (1987) summarized the karyological literature of plecotines and showed that Barbastella had a karyotype identical to several members of the plecotine group, further supporting its position as the nearest relative of the group. According to their scheme, however, Barbastella was part of a monophyletic group that also included Plecotus and Corynorhinus, but excluded Euderma and Idionycteris. Hill and Harrison (1987) have suggested, on the basis ofgeneral trends in bacular morphology, that more proximate relatives to the traditional plecotine group are Rhogeessa, Baeodon (usually considered part of Rhogeessa), Nycticeius, and Otonycteris, with the myotine group next closest. Our concerns about this arrangement are: (1) The bacular types of Hill and Harrison are sufficiently subjective that, beyond a general sense of "plecotine-ness," we could not decide from their illustrations and text what characteristics would define a plecotine baculum, although the traditionally viewed plecotine group (including Barbastella) does form a morphologically compact cluster. At least some taxa (e.g., Eptesicus nasutus and Bauerus dubiaquercus), which are excluded from the plecotines by Hill and Harrison, appear to us from their illustrations to have bacula very similar to those of plecotines. Without a more rigorous definition, this general morphology is unlikely to be particularly informative at this level of universality. (2) Without any discussion of transformational polarity, their taxonomic suggestions rest on overall similarity. The plecotine-type baculum (even ifit could be defined adequately) could easily be plesiomorphic within the vespertilionids, in which case the Similarities seen by Hill and Harrison would be phylogenetically uninformative. (3) If the plecotines (sensu Hill and Har-

4 4 AMERICAN MUSEUM NOVITATES NO rison) form a monophyletic group (that is, if Rhogeessa, Baeodon, Nycticeius, and Otonycteris are plecotines) then the karyological evidence of Bickham (1979) as well as much of the osteological and dentitional evidence presented by Tate (1942) would be phylogenetically incoherent. That Hill and Harrison discounted dentitional evidence as "adaptive" (p. 260) is insufficient; convergence in dentition can only be documented on the basis of other evidence, not merely by asserting that convergence is plausible, just because convergence has been documented elsewhere. Additionally, the congruence ofthe evidence of Bickham (1979) and Tate (1942) should not be so easily dismissed. If one takes the trees of Bickham (1979) and Tate (1942) to be undirected diagrams ofrelationship, considerable consensus exists between them. That is, in both studies Rhogeessa, Baeodon (not reported by Bickham, but usually considered very close to Rhogeessa), Nycticeius, and Otonycteris (also not reported by Bickham, but also usually considered to be related to Nycticeius) are patristically distant from the traditionally viewed plecotines, and the plecotines are patristically close to Lasiusus, Lasionycteris, and "Myotis," which has no unambiguous apomorphies with respect to the plecotines. We have therefore set aside Hill and Harrison's (1987) view as based on a morphological system (bacula) whose "window of applicability" would seem to be at a considerably lower level of universality than the level at which it was applied. We have therefore used "Myotis" (composed in our analysis of "M." liebii, "M." lucifugus, "M." myotis, "M." sodalis, and "M." thysanodes) as a functional outgroup on which to root our plecotine tree. MATERIALS AND METHODS Using information from the literature (e.g., Tate, 1942; Handley, 1959), augmented and verified by our own observations, we recognized 25 transformation series of morphological characters for each species. Additionally, 12 transformation series of karyology were derived from the literature (Bickham, 1979; Leniec et al., 1987; Stock, 1983). Outgroup comparison has been selected as the most general means of polarizing character transformation series (Watrous and Wheeler, 1981; Farris, 1982; Kluge, 1985; Brooks and Wiley, 1985). These independently polarized transformation series were analyzed by using the branch-and-bound algorithm (Hendy and Penny, 1982) included in the PAUP (Swofford, 1989) and Hennig86 (Farris, 1988) programs under both (1) the assumption of additivity (i.e., multicharacter morphoclines allowed) and (2) not under the assumption of additivity (i.e., multicharacter morphoclines not allowed). This allowed some evaluation of the distortion caused by the assumption of character additivity, rather than simple character matching. Tree fit was evaluated with the consistency index (ci) of Kluge and Farris (1969). In this measure, a value of "l" means that all data are logically consistent with all other data on the tree; decreasing values reflect an increase in homoplasy. Only those features that were placed unambiguously, regardless of the character optimization methodology employed (i.e., accelerated transformation or delayed transformation) or of assumptions about character additivity, were considered to be evidence of relationship. We have not restricted ourselves to a particular "kind" ofdata (e.g., morphological vs. karyological) because it is incumbent on us to explain all the evidence at hand (Miyamoto, 1985; Kluge, 1989). Nor have we employed differential character weighting. This practice has been argued against convincingly by Patterson (1982) and Novacek (1986). In short, we think that differential character weighting removes much of the objectivity from discussions of character incongruence. One multicharacter transformation series was used (Number 16 below) that could not be polarized by appeal to the functional outgroup and was treated as nonadditive even in the additive analysis. Specimens examined are listed in Appendix 1. TRANSFORMATION SERIES The source of the transformation series, if originally gleaned from the literature, is cited. Under the assumption of character additivity, "0" is the plesiomorphic condition, and "1" (and higher integers) represent succes-

5 1992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 5 sively derived conditions. In the nonadditive analysis, or Transformation Series 16 in the additive analysis, the distance between alternative characters is considered to be 1, regardless of the integer assignments of the characters. Although some other features (e.g., calcar development [Handley, 1959], attachment point of interfemoral membrane [Handley, 1959], expansion of acromion process and notching ofmanubrium (= presternum) [Hall, 1934]) have been suggested to reflect phylogenetic information, they have not been included in our list oftransformation series because ofintraspecific variation or our inability to verify early observations. Synapomorphies of the ingroup (discussed above) and autapomorphies of terminal taxa (e.g., the pink ears and dorsal white spots ofeuderma maculatum) are excluded from the analysis for clarity of discussion. Transformation series listed below are reflected in the data matrix of Appendix 2. Original sources for each character description are provided. 1. Rostrum (Handley, 1959): (0) flattened, with median concavity; (1) flattened, with slight concavity; (2) arched, without median concavity. 2. Braincase (Handley, 1959): (0) relatively shallow; dorsal surface of skull relatively flat; (1) relatively deep-doming of skull prevents the skull from lying flat while on its dorsal surface. 3. Preorbital-supraorbital region (Handley, 1959): (0) smoothly rounded or faintly ridged; (1) sharply ridged. 4. Temporal ridges (Handley, 1959): (0) confluent medially interorbitally (or nearly so); (1) not confluent medially, with distinct muscle scars. Idionycteris and Euderma have the temporal ridges completely separated. This condition is approached in Plecotus, in which the temporal ridges meet at a point interorbitally but separate posterior to this point. In Corynorhinus the temporal ridges meet interorbitally and do not separate posterior to this point. Although we could have cast this transformation into three characters, outgroup comparison difficulties in distinguishing the Plecotus and Corynorhinus conditions required us to recognize only the Euderma-Idionycteris condition as apomorphic. 5. Zygoma, fragility (Handley, 1959): (0) relatively thin and fragile; (1) relatively thick and strong. 6. Zygoma, postorbital expansion (Handley, 1959): (0) postorbital expansion in middle or slightly anterior to middle ofzygomatic arch, or not expanded; (1) postorbital expansion in posterior third of arch. 7. Auditory bullae, shape (Handley, 1959): (0) roughly circular in outline, slightly enlarged; (1) enlarged, slightly elongate; (2) enlarged and elliptical. 8. First upper incisors (I I and 12) (Handley, 1959): (0) I2 and II subequal; (1) II much taller than I2 (12 reduced). 9. Upper fourth premolar (P4) (Handley, 1959): (0) much wider than long, posterolingual part of tooth not reduced; (1) approximately equal in length and width, posterolingual part of tooth reduced. 10. Metacone of upper third molar (M3) (Handley, 1959): (0) present; (1) absent Anterointemal cusp oflower canine (cl) (Handley, 1959): (0) relatively large, ca size of primary cusp in height; (1) small and greatly exceeded by primary cusp. This feature reflects the absolute size of the secondary cusp. 12. Lower third premolar (p3) (Handley, 1959): (0) cross-sectional outline not distorted; (1) cross-sectional outline distorted. Because Barbastella and Euderma lack a p3, they are coded as unknown for this feature. 13. Lower third premolar (p3) (Handley, 1959): (0) present; (1) absent. 14. Lower fourth premolar (p4) (Handley, 1959): (0) double-rooted; (1) single-rooted. Handley (1959: 11) reported Corynorhinus townsendii as having only one root on the p4. However, all Corynorhinus we observed (including townsendii) had two roots, although these were very small and in some cases sometimes appeared to be single. 15. Posterior extension of anterior naris (Handley, 1959): (0) vomers not exposed; (1) posterior extension of anterior naris so far back as to expose vomers. 16. Median postpalatal process (Handley, 1959): (0) weak single spine; (1) prominent single spine; (2) bifid prominence; (3) absent. Because of outgroup ambiguity, this must be treated as a nonadditive (= unordered) set of

6 6 AMERICAN MUSEUM NOVITATES NO characters, even though condition 0 (weak single spine) clearly is the ancestral condition. 17. Basial (= basicranial) pits (Handley, 1959): (0) absent; (1) present. 18. Manubrium (= prestemum of Hall, 1934; Handley, 1959): (0) as wide or wider than long; (1) distinctly longer than wide. 19. Second phalanx ofthird digit (Handley, 1959): (0) less than or equal in length to first phalanx; (1) much longer than first phalanx. 20. Nostril (Handley, 1959): (0) unspecialized; (1) great posterior elongation with a conspicuous shallow basin posteriad and separated by a septum. 21. Muzzle glands (Handley, 1959): (0) absent; (1) present, not greatly enlarged; (2) greatly enlarged. 22. Ears (Dobson, 1875; Handley, 1959): (0) small; (1) extremely enlarged. 23. Tragus (Handley, 1959): (0) narrow, bladelike; no prominent constriction near the base; (1) intermediate; (2) paddlelike, with a prominent constriction near the base. 24. Accessory anterior basal lobe ofauricle (Handley, 1959): (0) absent; (1) present (weak in Plecotus: a projecting lappet in Idionycteris). 25. Transverse ribs on auricle (Handley, 1959): (0) uninterrupted by vertical rib near posterior border; (1) interrupted by vertical rib near posterior border. 26. Chromosome arms 21 and 7 (Bickham, 1979; Leniec et al., 1987): (0) unfused; (1) 27. Chromosome arms 13 and 10 (Stock, 28. Chromosome arms 21 and 9 (Stock, 29. Chromosome arms 19 and 8 (Stock, 30. Chromosome arms 22 and 12 (Stock, 31. Chromosome arms 18 and 14 (Stock, 32. Chromosome arms 23 and 20 (Stock, 33. Chromosome arms 19 and 14 (Stock, 34. Chromosome arms 22 and 8 (Stock, 35. Chromosome arms 13 and 9 (Stock, 36. Chromosome arms 12 and 10 (Stock, RESULTS Statistics presented below, unless bracketed, refer to the analysis in which additivity of multicharacter transformations was assumed. Bracketed statistics refer to the data analysis in which all character matches were regarded as nonadditive (= unordered). Analysis of the data matrix (Appendix 2), with or without the assumption of transformational additivity, produced one most parsimonious tree (fig. 1, Appendices 3 and 4), which has a length of 53 [50] (ci = [0.840]), and shows Barbastella within the plecotines, as the sister taxon of Plecotus + Corynorhinus, a result in accord with those of Leniec et al. (1987). Because the karyological data were taken directly from the literature and were not our observations we also analyzed a matrix containing only morphological data (Transformation Series 1-25) that we could verify. Under the assumption of additivity this matrix produced three trees (length = 41, ci = 0.756), all of which (fig. 2; trees 1-3) show considerably different topologies from that based on the entire data set. Without the assumption of additivity these same three trees were discovered (length = 39; ci = 0.795), with the addition ofone more of the same length (fig. 2; tree 4). However, if the complete data set (Transformation Series 1-36) is forced onto trees 1-3 (fig. 2), the resulting statistics are length = 57 [55], ci = [0.764]. In other words, the topologies based solely on morphology require, when assuming additivity, four more steps in the entire data set, and five more steps when additivity is not assumed, than does the most parsimonious topology based on the entire data set. Tree 4 (fig. 2), discovered by the

7 1 992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 7 6 Fpsl baei2 Fig. 1. Most parsimonious tree for all data. Numbered stems are documented in Appendices 3 and 4. morphology-only data set, when additivity is not assumed, is identical to the topology discovered by the entire data set (fig. 1), both with additivity and without (length = 53 [50]; ci = [0.840]). In other words, without additivity assumed, the morphological transformations do not reject the tree that is most parsimonious for all of the data, regardless of the assumptions made about additivity of character matches. For those trees that are supported under the assumption of additivity, the morphological evidence is only marginally incongruent with the karyological data, regardless of the most parsimonious trees produced solely by morphology. When additivity is assumed, the restricted morphological data set cast onto the topology produced by the entire data set has a length of 42 (ci = 0.738), which is only one step longer than the trees (fig. 2, trees 1-3) based solely on the morphological data set (length = 41, ci = 0.756). This means that regardless of the relatively high consistency indices produced by the morphology-only analyses, it is clear that the morphology-only data set is only marginally more supportive ofits peculiar trees (fig. 2, trees 1-3) than that tree that is most parsimonious ofall data (fig. 1 and fig. 2, tree 4). This is a good example of why consistency indices should not be taken as measures of tree stability. We hesitate to make any generalizations about character additivity, inasmuch as we think that this particular issue is inextricably joined to the issue of how similar character matches must be for the systematist to hypothesize homology. However, in this particular case, the only serious departure from congruence between the additive and nonadditive analysis is in the placement of characters in Transformation Series 7 (bullae shape and size). Although we accept the tree most supported by the complete data set, this means that we must accept either that enlarged ears of Euderma + Idionycteris are not homologous with those in Corynorhinus + Plecotus, or that Barbastella has, as a synapomorphy

8 8 AMERICAN MUSEUM NOVITATES NO Fig. 2. \/4 \/ Alternative trees produced solely by the morphological subset of the data. of that genus, reduced the size of the ears from "enlarged" to "normal." Although we suspect that Barbastella has reduced its ear size, this issue cannot be resolved by this data set Ḋata supporting the stems noted in figure 1 are noted in Appendices 3 and 4. COMPARISON WITH PREVIOUS VIEWS Only Leniec et al. (1987) have suggested that Barbastella is within the traditional plecotines, rather than the sister taxon of traditional plecotines. Because a large proportion of the data here analyzed are theirs, it is not surprising that we came to identical conclusions. However, as discussed above, the morphological data are only marginally at variance with the karyological evidence. Excluding the placement ofbarbastella, our arrangement is most similar to that proposed by Williams et al. (1970), but contrasts with that of Handley (1959). The tree suggested by Handley (1959) (fig. 3) fared particularly poorly (length = 68 [65], ci = 0.62 [0.65]), when the entire data set is cast upon it, requiring an additional 16 convergences under the assumption of additivity and 13 without this assumption. This is not surprising, because Handley (1959) based his view on an evaluation ofoverall similarity. The tree suggested by Williams et al. (1970) (fig. 4) (identical to our Tree 3 in fig. 3 for the morphology-only data set) fared better (length = 57 [55], ci = 0.74 [0.76]), largely because of the placement of Idionycteris as the sister taxon of Euderma rather than the sister taxon of Corynorhinus + Plecotus as posited by Handley. SELECTION OF A TAXONOMY A taxonomy should be logically consistent with that which it purports to document (Hull, 1964; Wiley, 1981), which, as most system-

9 1 992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 9 Fig. 3. Phylogeny posited by Handley (1959). atists now agree, is recovered phylogenetic history. A rather large number of alternative taxonomies exist that are logically consistent with our preferred cladogram (fig. 1) (e.g., one genus for all ingroup species, or different genera for each species in the ingroup). However, we think that a taxonomy should maximize phylogenetic retrievability without undue violence to the traditional taxonomy or to perturbability from future discoveries. For this reason we recommend the following taxonomy: Tribe Plecotini Dobson, 1875 Euderma H. Allen, 1892 E. maculatum (J. A. Allen, 1891) E. phyllote (G. M. Allen, 1916)4 Barbastella Gray, 1821 B. barbastella (Schreber, 1774) B. leucomelas (Cretzschmar, 1826) Plecotus Geoffroy Saint-Hilaire, 1818 P. auritus (Linnaeus, 1758) P. austriacus (Fischer, 1829) P. teneriffae Barrett-Hamilton, Corynorhinus H. Allen, 1865 C. mexicanus G. M. Allen, 1916 C. rafinesquii (Lesson, 1827) C. townsendii (Cooper, 1837) Idionycteris is synonymized with Euderma because they are sister species with few autapomorphies. We could have retained Corynorhinus as a subgenus of Plecotus, but this taxonomy is dependent on a cladogram whose arrangement is only slightly more parsimo- 4 The original formation of the species name as an adjective rather than a noun makes for this rather unhappy spelling. I Provisionally allocated here; specimens not seen. Fig. 4. Phylogeny posited by Williams et al. (1970). nious than trenchantly different arrangements that would not place Corynorhinus and Plecotus as sister taxa (e.g., considering Barbastella and Corynorhinus to be sister taxa requires only one additional step in the overall tree assuming additivity and only two more steps without this assumption). TAXONOMIC ACCOUNTS With the exception of the diagnosis for the Plecotini which lists all known unambiguous synapomorphies, the intention of the diagnoses below is to provide a list of morphological characteristics for identifying named taxa. Unambiguous synapomorphies of diagnosed groups are italicized. Unnamed taxa are diagnosed in Appendix 4. TRIBE PLECOTINI Plecoti Dobson, 1875: 349. Type genus: Plecotus Geoffroy Saint-Hilaire, DIAGNosIs: Zygomatic arches do not extendfarfrom the skull;,fusion ofchromosome arms 15 and 11; ears connected by an interauricular septum (low in Corynorhinus). DISTRIBUTION: Temperate Eurasia, North Africa, and North America. Genus Euderma Euderma H. Allen, 1892 [1891]: 467. Type species: Histiotus maculatus J. A. Allen, Idionycteris Anthony, 1923: 1. Type species: Idionycteris mexicanus Anthony, 1923 (= Corynorhinus phyllotis G. M. Allen, 1916). DIAGNOSIS: Rostrum not flattened, with median concavity (flattened with slight concavity in Plecotus and arched with median

10 10 AMERICAN MUSEUM NOVITATES NO concavity in Corynorhinus); braincase dorsally flattened as in Plecotus (domed in Corynorhinus and Barbastella); zygoma with postorbital expansion in middle or slightly anterior to middle of zygomatic arch (posterior expansion in posterior third of arch in Corynorhinus); auditory bullae very enlarged and elliptical (slightly enlarged, circular in Barbastella; enlarged and slightly elongate in Corynorhinus and Plecotus); P4 much wider than long (length and width subequal in Plecotus); metacone on M3 present (absent in Plecotus); p3 absent in E. maculatum, present in E. phyllote (also absent in Barbastella); p4 single rooted (double-rooted in Corynorhinus, Barbastella, and Plecotus); basial pits absent (present in Corynorhinus); ears extremely enlarged (small in Barbastella); tragus paddlelike, with a prominent constriction near the base (narrower in Plecotus, Barbastella, and Corynorhinus); muzzle glands not enlarged (weakly enlarged in Plecotus; very enlarged in Corynorhinus and Barbastella); transverse ribs on auricle uninterrupted by vertical rib near posterior border (interrupted in Corynorhinus). CONTENT: Euderma maculatum (J. A. Allen, 1891), and E. phyllote (G. M. Allen, 1916). DISTRmurIoN: Western North America from southwestern Canada to San Luis Potosi and Mexico City, Mexico. Genus Barbastella Barbastella Gray, 1821: 300. Type species: Ves- -pertilio barbastellus Schreber, Synotus Keyserling and Blasius, 1839: 305. Type species: Vespertilio barbastellus Schreber, DLAGNosIs: Rostrum flattened, with median concavity (with slight concavity in Plecotus and arched in Corynorhinus); braincase dorsally domed as in Corynorhinus (flattened in Euderma and Plecotus); zygomatic arch with expansion near middle (in posterior third in Corynorhinus); auditory bullae slightly enlarged, circular in outline (enlarged in Plecotus, Corynorhinus, and Euderma); P4 much wider than long (subequal in Plecotus); metacone of M3 present (absent in Plecotus); p3 absent (present in Euderma phyllote, Plecotus, and Corynorhinus); p4 double rooted (single rooted in Euderma); basial pits absent (present in Corynorhinus); ears of moderate size (extremely enlarged in Euderma, Corynorhinus, and Plecotus); tragus narrow (wider in Plecotus and Euderma); muzzle glands greatly enlarged; transverse ribs on auricle not interrupted posteriorly by vertical rib on posterior border (interrupted in Corynorhinus). CONTENT: Barbastella barbastella (Schreber, 1774), and B. leucomelas (Cretzschmar, 1826). DISTRIBUTION: Western Europe, Morocco, and Ethiopia to Caucasus, Pamirs, and western China; Japan. Genus Plecotus Plecotus Geoffroy Saint-Hilaire, 1818: 212. Type species: Vespertilio auritus Linnaeus, DiAGNOSIS: Rostrum flattened, with slight concavity (convexly arched in Corynorhinus, larger concavity in Euderma); braincase dorsally flattened as in Euderma (domed in Corynorhinus and Barbastella); zygomatic arch with expansion near middle (in posterior third in Corynorhinus); auditory bullae enlarged, slightly elongate (smaller in Barbastella; much larger in Euderma); P4 subequal in length and width (much wider than long in Euderma, Barbastella, and Corynorhinus); metacone of M3 absent (present in Euderma, Barbastella, and Corynorhinus); p3 present (absent in Barbastella and Euderma maculatum); p4 double rooted (single rooted in Euderma); basial pits absent (present in Corynorhinus); ears very large (smaller in Barbastella); tragus moderately broad (very broad in Euderma; narrow in Corynorhinus and Barbastella); muzzle glands visible but not greatly enlarged (large in Corynorhinus and Barbastella); transverse ribs on auricle not interrupted posteriorly by vertical rib on posterior border (interrupted in Corynorhinus). CONTENT: Plecotus auritus (Linnaeus, 1758), P. austriacus (Fischer, 1829), and P. teneriffae Barrett-Hamilton, DIsTRIBUrION: Norway, England, Spain, and Senegal to Mongolia, Japan; Canary and Cape Verde Islands. Genus Corynorhinus Corynorhinus H. Allen, 1865: 173. Type species: Plecotus macrotis Le Conte, 1831 (= Plecotus rafinesquiilesson, 1827).

11 1992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS I1I DIAGNOSIS: Rostrum arched, without median concavity (flattened with varying degrees of development of median concavity in Euderma, Barbastella, and Plecotus); braincase dorsally domed as in Barbastella (flattened in Euderma and Plecotus; expansion of zygomatic arch in posterior third (near middle in Euderma, Barbastella, and Plecotus); auditory bullae enlarged, slightly elongate (larger, more elongate in Euderma; less enlarged in Barbastella); P4 much wider than long (subequal in Plecotus); metacone of M3 present (absent in Plecotus); p3 present (absent in Barbastella and Euderma phyllote); p4 double rooted (single rooted in Euderma); basial pits present (absent in Euderma, Barbastella, and Plecotus); ears very large (smaller in Barbastella); tragus narrow, bladelike (broader in Plecotus and Euderma); muzzle glands greatly enlarged (smaller in Euderma, Barbastella, and Plecotus); transverse ribs on auricle interrupted by vertical rib near posterior border (uninterrupted in Euderma, Plecotus, and Barbastella). CONTENT: Corynorhinus mexicanus G. M. Allen, 1916, C. rafinesquii (Lesson, 1827), and C. townsendii (Cooper, 1837). DISTRmUTION: Southern British Columbia (Canada) and Virginia (USA) south to southeastern Mexico and the Isthmus of Tehuantepec. REFERENCES Allen, G. M Bats of the genus Corynorhinus. Bull. Mus. Comp. Zool. 60: Allen, H On a new genus of Vespertilionidae. Proc. Acad. Nat. Sci. Philadelphia 17: [ A new genus of Vespertilionidae. Ibid., 43: Allen, J. A Description ofa new species ofbig-eared bat, of the genus Histiotus, from southern California. Bull. Am. Mus. Nat. Hist. 43: x pp. Anthony, H. E Mammals from Mexico and South America. Am. Mus. Novitates 54: 10 pp- Baker, R. J., B. L. Davis, R. G. Jordan, and A. Binous Karyotypic and morphometric studies oftunisian mammals: Bats. Mammalia 38: Barrett-Hamilton, G. E. H Descriptions of two new species of Plecotus. Ann. Mag. Nat. Hist. ser. 7, 20: Bickham, J. W Chromosomal variation and evolutionary relationships of vespertilionid bats. J. Mammal. 60: Brooks, D. R., and E. 0. Wiley Theories and methods in different approaches to phylogenetic systematics. Cladistics 1: Cooper, W On two species of Plecotus inhabiting the United States Territory. Ann. Lyc. Nat. Hist. 4: Cretzschmar, P. J Saugethiere. In W. P. E. S. Rueppell, Atlas zu der Reise in nordlichen Africa,... Erste Abtheilung, Zoologie. Hrsg. von der Senkenbergischen naturforschenden Gesellschaft. Frankfurt am Main: H. L. Bronner. Dobson, G. E Conspectus of the suborders, families, and genera of Chiroptera arranged according to their natural affinities. Ann. Mag. Nat. Hist. ser. 4, 16: Farris, J. S Outgroups and parsimony. Syst. Zool. 31: Hennig86. Version 1.5. Computer program distributed by J. S. Farris, Port Jefferson, New York. Fischer, J. B Synopsis mammalium. Stuttgardtiae: J. G. Cottae, xlii pp. Geoffroy Saint-Hilaire, E Description des Mammiferes qui se trouvent en Egypte. In Vol. 2 (Histoire Naturelle), pp In Description de l'egypte, ou Recueil des Observations et des Recherches qui ont ete faites en Egypte pendant l'expedition de l'armee Fran9aise, publie par les Ordres de sa Majeste l'empereur Napoleon le Grand. Paris. Gray, J. E On the natural arrangement of vertebrose animals. London Med. Reposit. 15: Hall, E. R Certain osteological features of Euderma maculatum. J. Mammal. 15: Handley, C. O., Jr A revision of American bats of the gen-

12 12 AMERICAN MUSEUM NOVITATES NO era Euderma and Plecotus. Proc. U.S. Natl. Mus. 110: Hendy, M. D., and D. Penny Branch and bound algorithms to determine minimal evolutionary trees. Math. Biosci. 59: Hill, J. E., and D. L. Harrison The baculum in the Vespertilioninae (Chiroptera: Vespertilionidae) with a systematic review, a synopsis of Pipistrellus and Eptesicus, and the descriptions ofa new genus and subgenus. Bull. Br. Mus. (Nat. Hist.), Zool. 52: Honacki, J. H., K. E. Kinman, and J. W. Koeppl (eds.) Mammal species of the world: A taxonomic and geographic reference. Lawrence, KS: Allen Press and Assoc. System. Collect., ix pp. Hull, D. L Consistency and monophyly. Syst. Zool. 13: Ib'afiez, C., and R. Fernandez 1986 [1985]. Systematic status ofthe long-eared bat Plecotus teneriffae Barret-Hamilton [sic], 1907 (Chiroptera; Vespertilionidae). Siiugetierkd. Mitt. 32: Keyserling, A., Graf von, and J. H. Blasius Uebersicht der Gattungs- und Artcharaktere der europiiischen Fledermause. Arch. Naturgesch. 5, 1: Kluge, A. G Ontogeny and phylogenetic systematics. Cladistics 1: A concern for evidence and a phylogenetic hypothesis of relationships among Epicrates (Boidae, Serpentes). Syst. Zool. 38: Kluge, A. G., and J. S. Farris Quantitative phyletics and the evolution of anurans. Syst. Zool. 18: Koopman, K. F. 1984a. Bats. In S. Anderson and J. K. Jones, Jr. (eds.), Orders and families of Recent mammals of the world, pp New York: Wiley. 1984b. A synopsis of the families of bats-part VII. Bat Res. News 25: Le Conte, J. L p. 433 ("Appendix II of the American editor"). In G. J. L. N. F. Cuvier, The Animal Kingdom, arranged in conformity with its organisation. [Translated from the French with notes and additions by H. McMurtrie.] New York: G. & C. & H. Carvill, 1: xxxi pp. Leniec, H., S. Fedyk, and A. L. Ruprecht Chromosomes of some species of vespertilionid bats IV. New data on the plecotine bats. Acta Theriol. 32: Lesson, R.-P Manuel de mammalogie, ou Histoire naturelle des mammiferes. Paris: Roret, xvi pp. Linnaeus, C Systema Naturae per Regna tria Naturae, secundum Classes, Ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, Locis. 10th ed., vol. 1. Stockholm. Miller, G. S., Jr Revision ofthe North American bats of the family Vespertilionidae. N. Am. Fauna 13: The families and genera of bats. Bull. U.S. Natl. Mus. 57: xvii pp. Miyamoto, M. M Consensus cladograms and general classifications. Cladistics 1: Nelson, G Classification as an expression of phylogenetic relationships. Syst. Zool. 22: Novacek, M. J The skull of leptictid insectivorans and the higher-level classification ofeutherian mammals. Bull. Am. Mus. Nat. Hist. 183: Patterson, C Morphological characters and homology. In K. A. Joysey and A. E. Friday (eds.), Problems of phylogenetic reconstruction, pp London: Academic Press. Schreber, J. C. D. von Die Siiugthiere in Abbildungen nach der Natur, mit Bescreibungen. Fortgesetzt von d. August Goldfuss... Erlangen in der Expedition des Schreber'schen Siiugthier- und des Esper'schen Schmetterlingswerkes, vol. 1. (Der Mensch. Der Affe. Der Maki. Die Fledermaus): iv pp. Simpson, G. G The principles of classification and a classification of mammals. Bull. Am. Mus. Nat. Hist. 85: i-xvi, Sokolov, V. E Sistematika Mlekopitayushchikh [Mammal Systematics]. Izdel'stvo Vysshaya Shkola: Moscow. Stock, A. D Chromosomal homologies and phylogenetic relationships of the vespertilio-

13 1992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 13 nid bat genera Euderma, Idionycteris, and Plecotus. Cytogenet. Cell Genet. 35: Swofford, D. L PAUP: Phylogenetic Analysis Using Parsimony. Version 3.01q. Computer program distributed by the Illinois Natural History Survey, Champaign, I1. Tate, G. H. H Results of the Archbold Expeditions. No. 47. Review of the vespertilionine bats, with special attention to genera and species of the Archbold collections. Bull. Am. Mus. Nat. Hist. 80: Watrous, L. E., and Q. D. Wheeler The out-group comparison method of character analysis. Syst. Zool. 30: 11 pp. Wiley, E An annotated Linnaean hierarchy, with comments on natural taxa and competing systems. Syst. Zool. 28: Convex groups and consistent classifications. Syst. Bot. 6: Williams, D. F., J. D. Druecker, and H. L. Black The karyotype of Euderma maculatum and comments on the evolution of the plecotine bats. J. Mammal. 51: APPENDIX 1 Specimens Examined Unless noted otherwise, specimens are skin and skull. AMNH = American Museum of Natural History; FMNH = Field Museum ofnatural History; KU = Museum of Natural History, University of Kansas; USNM = National Museum of Natulral History, Smithsonian Institution. OUTGROUP TAXA: "Myotis" liebii: KU (skeleton). "M." lucifugus: AMNH (alcoholic); KU , 45536, (skeleton). "M." myotis: AMNH (alcoholic). "M." sodalis: KU (skeleton). "M." thysanodes: KU , INGROUP TAXA: Barbastella barbastella:amnh (alcoholics; x-rays), (skull); KU 88261; USNM 37568, (alcoholic; x-ray), , B. leucomelas: AMNH (skull); FMNH (alcoholic; x-ray), ; KU Corynorhinus mexicanus: KU (skeleton), (alcoholics), (skeletons), (alcoholics), (alcoholic), (alcoholics); USNM (alcoholics). C. rafinesquii: KU 11183, , 71989, 99392; USNM (alcoholic; x-ray), (alcoholic, x-ray). C. townsendii: AMNH (alcoholics), (skeletons), (skeletons), (skeleton), (alcoholic); FMNH (partial skeleton); KU 5210 (skeleton), (skeletons), (alcoholic), (alcoholics); USNM (alcoholic, x-ray), (alcoholic, x-ray). Euderma maculatum: FMNH (skeleton); KU (skin and skeleton), (skin, skull, and alcoholic body). E. phyllote: AMNH (skeleton); FMNH (partial skeleton); KU 73211, (skeleton), (skeletons), (skeleton), , ; USNM (alcoholic; x-ray). Plecotus auritus: AMNH (alcoholic), (skull), (skull), (alcoholic; x-ray); FMNH 91634; KU (skull), (skull), (alcoholic), 88262; USNM (skeleton), P. austriacus: AMNH (alcoholic; x-ray); FMNH ; KU

14 14 AMERICAN MUSEUM NOVITATES NO APPENDIX 2 Data Matrix Hypothetical Ancestor Barbastella leucomelas ? ??????????? Barbastella barbastella ? Euderma maculatum O?llO Idionycteris phyllotis Plecotus auritus Plecotus austriacus Corynorhinus mexicanus ??????????? Corynorhinus rafinesquii ??????????? Corynorhinus townsendi

15 1 992 FROST AND TIMM: PHYLOGENY OF PLECOTINE BATS 15 APPENDIX 3 Changes, by Transformation Series, for the Stems Noted on Figure 1 Trans. series ci Steps Change Along stem to t :t #1I -O to tl r [ # f1-0 E. maculatum E. maculatum f2-3 I. phyllotis APPENDIX 3-(Continued) Trans. series ci Steps Change Along stem t # # L phyllotis ' t Denotes a change placed on that stem under only one character transformation optimization (e.g., accelerated transformation in this case) under the assumption of character additivity. t Denotes ambiguous placement due to character optimization alternatives under no assumption ofcharacter additivity. For transformation series 7, the alternative arrangement in brackets refer to results under no assumption of additivity.

16 16 AMERICAN MUSEUM NOVITATES NO APPENDIX 4 Apomorphy Lists by Stem from Figure 1 Trans. Stem series Steps ci Change 1 ( to - 1) f0-1l :to #0 - I tl to - I [ ] b #Il t t I to -l APPENDIX 4-(Continued) Trans. Stem series Steps ci Change ( ) [ ] t b b b ' I E. macu $l - o latum I. phyllotis f I1 t Denotes a transformation whose placement is dependent on a particular character optimization (e.g., accelerated transformation) under an assumption of character additivity. t Denotes a transformation whose placement is dependent on a particular character optimization under no assumption of additivity. Transformations in brackets are only in the results of the nonadditive analysis; those in parentheses are only in the additive analysis. Recent issues of the Novitates may be purchased from the Museum. Lists of back issues of the Novitates, Bulletin, and Anthropological Papers published during the last five years are available free of charge. Address orders to: American Museum of Natural History Library, Department D, Central Park West at 79th St., New York, N.Y I THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER.